Skip to content

Using ICD-10 and CCAM

eds-scikit provides two functions to ease the extraction of occurrrences of

  • ICD-10 codes : eds_scikit.event.icd10.conditions_from_icd10
  • CCAM codes : eds_scikit.event.ccam.procedures_from_ccam

These two functions are by design similar. In fact, they call under the hood the same base function.

Let us see a minimal working example that would allow us to select patients with Deep Vein Thrombosis based on the presence of specific ICD-10 codes.

from eds_scikit.io import HiveData
data = HiveData(DBNAME)

codes = dict(
    DVT = dict(
        exact = ["I81", "O223", "O082", "O871"], # (1)
        regex = ["I82[02389]", "I80[12]"]
    )
)

from eds_scikit.event.icd10 import conditions_from_icd10

DVTs = conditions_from_icd10(
    condition_occurrence=data.condition_occurrence,
    visit_occurrence=data.visit_occurrence,
    codes=codes,
    date_from_visit=True,
    additional_filtering=dict(
        condition_status_source_value={"DP", "DAS"}, # (1)
    )
)
  1. Here you can provide either exact, regex or prefix codes
  2. With this syntax we will keep only DP (Diagnostic Principal) or DAS (Diagnostic Associé) diagnoses

Of course, you are encouraged to check the documentation of those functions as they provide additional parameters that might be useful depending on your needs.